Computer-Assisted Segmentation of Craniofacial Structures and Osteotomy Planning Using CBCT and CT
Development and Evaluation of Methods for Automatic and Semi-Automatic Segmentation of Craniofacial Structures and Delineation of Osteotomy Corridors Using CBCT and CT Imaging
This project will develop and evaluate methods for automatic and semi-automatic segmentation of craniofacial structures and delineation of osteotomy corridors using cone-beam computed tomography (CBCT) and computed tomography (CT) images. It will assess the quality and reproducibility of expert-prepared reference segmentations, the influence of model training, and the accuracy, processing time, and errors of automated methods. Only existing clinical scans will be used, with no additional imaging or changes to patient treatment.
調査の概要
状態
詳細な説明
This retrospective methodological project will use pseudonymized CBCT and CT images. Expert-reviewed segmentations and virtual osteotomy plans will serve as references. Selected cases will undergo repeated and independent assessment to evaluate intra- and interobserver agreement. A development dataset of CBCT examinations will be used to assess the effects of training set size and training progression. The final model will be tested on examinations not used during development. Evaluation will include segmentation accuracy, processing time, errors, technical failures, and correction requirements. Osteotomy corridors will be assessed against reference plans and nearby critical anatomical structures.
研究の種類
入学 (推定)
連絡先と場所
研究連絡先
- 名前：Maciej Sikora
- 電話番号：+48 41 260 55 85
- メール：sikora-maciej@wp.pl
研究場所
-
-
-
Kielce、ポーランド、25-375
- Department of Maxillofacial Surgery, Hospital of the Ministry of Interior
-
コンタクト：
- Maciej Sikora, dr hab.
- 電話番号：+48 41 260 55 85
- メール：sikora-maciej@wp.pl
-
-
参加基準
適格基準
就学可能な年齢
健康ボランティアの受け入れ
受講資格のある性別
サンプリング方法
調査対象母集団
Patients aged 16-85 years at the time of clinically indicated CBCT or CT imaging, whose archived examinations meet the study eligibility criteria.
説明
Inclusion Criteria:
- Archived CBCT or CT examinations acquired for clinical indications.
- Patient age of 16-85 years at the time of imaging.
- Images available in DICOM format and suitable for analysis.
- Complete visualization of the anatomical structures required for the planned analysis.
- Image quality sufficient for reliable reference preparation and assessment.
Exclusion Criteria:
- Incomplete or corrupted image series.
- Insufficient imaging coverage.
- Image quality preventing reliable reference assessment.
研究計画
研究はどのように設計されていますか？
デザインの詳細
- 観測モデル：OTHER
- 時間の展望：RETROSPECTIVE
コホートと介入
グループ/コホート |
介入・治療 |
|---|---|
|
Patients with archived CBCT or CT examinations
Patients whose archived CBCT or CT examinations will be used to develop and evaluate automatic and semi-automatic craniofacial segmentation methods and virtual osteotomy planning. |
Automatic and semi-automatic segmentation of craniofacial structures and virtual delineation of osteotomy corridors using archived CBCT and CT images. |
この研究は何を測定していますか？
主要な結果の測定
結果測定 |
メジャーの説明 |
時間枠 |
|---|---|---|
|
Segmentation accuracy compared with expert-prepared reference segmentations
時間枠：At the time of final model evaluation, up to 6 months
|
Agreement between uncorrected algorithm-generated segmentation masks and expert-prepared reference masks, assessed using the Dice similarity coefficient (DSC). Scores range from 0 to 1, with higher values indicating greater overlap. |
At the time of final model evaluation, up to 6 months
|
二次結果の測定
結果測定 |
メジャーの説明 |
時間枠 |
|---|---|---|
|
Intraobserver and interobserver agreement of reference segmentations
時間枠：During initial, repeat, and independent reference segmentation assessments, up to 6 months
|
Agreement between repeated reference segmentations by the same investigator and independent reference segmentations by a second investigator, assessed using the Dice similarity coefficient (DSC). Scores range from 0 to 1, with higher values indicating greater overlap. |
During initial, repeat, and independent reference segmentation assessments, up to 6 months
|
|
Segmentation processing time
時間枠：During segmentation processing, up to 6 months
|
Time required to generate segmentations for the evaluated examinations. |
During segmentation processing, up to 6 months
|
|
Frequency of segmentation errors, technical failures, and correction requirements
時間枠：During segmentation processing and initial expert evaluation, up to 6 months
|
Frequency of segmentation errors, technical failures, restarts, and correction requirements, reported separately by category. |
During segmentation processing and initial expert evaluation, up to 6 months
|
|
Segmentation performance by training set size and training progression
時間枠：During model training and internal validation, up to 6 months
|
Dice similarity coefficient values across evaluated training set sizes and training stages during internal validation. |
During model training and internal validation, up to 6 months
|
|
Evaluation of virtual osteotomy corridors
時間枠：At the time of virtual osteotomy corridor evaluation, up to 6 months
|
Assessment of computer-generated osteotomy corridors against expert-prepared reference plans and in relation to nearby critical anatomical structures. |
At the time of virtual osteotomy corridor evaluation, up to 6 months
|
協力者と研究者
研究記録日
主要日程の研究
研究開始 (推定)
一次修了 (推定)
研究の完了 (推定)
試験登録日
最初に提出
QC基準を満たした最初の提出物
最初の投稿 (実際)
学習記録の更新
投稿された最後の更新 (実際)
QC基準を満たした最後の更新が送信されました
最終確認日
詳しくは
本研究に関する用語
その他の研究ID番号
- CT/2026/3
個々の参加者データ (IPD) の計画
個々の参加者データ (IPD) を共有する予定はありますか?
医薬品およびデバイス情報、研究文書
米国FDA規制医薬品の研究
米国FDA規制機器製品の研究
この情報は、Web サイト clinicaltrials.gov から変更なしで直接取得したものです。研究の詳細を変更、削除、または更新するリクエストがある場合は、register@clinicaltrials.gov。 までご連絡ください。 clinicaltrials.gov に変更が加えられるとすぐに、ウェブサイトでも自動的に更新されます。